1. Join Now

    AVForums.com uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.

Bass Management - large or small

Discussion in 'Home Cinema Speakers' started by Ian J, May 24, 2002.

Thread Status:
Not open for further replies.
  1. Ian J

    Ian J
    Banned

    Joined:
    Aug 6, 2001
    Messages:
    25,529
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    166
    Ratings:
    +4,906
    I have just read an interesting article in this month's edition of Hifi Choice about bass management which seems to be contrary to accepted wisdom on setting "large" or "small" on AV amps.

    The author suggests that bass management is a far from tranparent process as apart from the phase shift that accompanies digital filterering, the redirection often causes big changes in both the levels and distortion of the signal.

    Comparative tests were carried out on several mid price AV amps with the speakers set to small and large comparing the amount of distortion measured on the bass signal which showed very little added distortion if there was no additional LFE signal being fed to the subwoofer but considerably more (dependant on amp) if there was LFE in addition to the redirected bass signal.

    Two of the amps tested showed an increase in distortion of up to 12% with speakers switched to small and LFE signal present over the comparative distortion with the speakers switched to large.

    A further two of the amps tested showed that the subwoofer feed increasing by 8dB just by switching from small to large which would either make a mockery of A/B testing or clip the subwoofer.

    According the the author the moral of the story is to set speakers to large regardless of the size of the speakers and avoid any bass management thus avoiding any additional phase shifts, changes in level or unwanted increases in distortion.

    I am no techie and have just paraphrased the article as an interesting discussion point for the technical gurus to get their teeth into.
     
  2. Jase

    Jase
    Well-known Member

    Joined:
    Mar 16, 2001
    Messages:
    9,315
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    136
    Location:
    South Coast
    Ratings:
    +827
    Surely redirecting the bass to a subwoofer will reduce the load on the amp, decreasing distortion??

    If the distortion is increasing then I would suggest the amps bass management isnt doing its job properly.

    To avoid using bass management you´ll need full range speakers that have a frequency response of 20hz-20khz (or 0hz-20khz) and a subwoofer that plumbs those depths as well, 0hz-20hz, not many people got those and indeed there arent that many speakers around that can.

    Lets see what the Technical boys say.:)
     
  3. lmccauley

    lmccauley
    Well-known Member

    Joined:
    Oct 24, 2000
    Messages:
    3,295
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    86
    Location:
    Newcastle
    Ratings:
    +212
    Yeah, that "woosh" noise was the sound of Ian's post flying over my head. Hopefully the usual suspects will be able to pull themselves away from their young children, installation businesses, writing, etc and comment.

    Cheers,
    Liam
     
  4. stranger

    stranger
    Guest

    Products Owned:
    0
    Products Wanted:
    0
    Ratings:
    +0
    i read this as well and tried it, all it did for my setup was to lose a lot of bass information. i think it would depend on equipment and a million other things and his results on distortion are probably only measurable in the lab. its too general for me and sister mag home ent. is following the same lead. free upgrade- downgrade imo with most gear.
     
  5. Dubbing Mixer

    Dubbing Mixer
    Standard Member

    Joined:
    Jan 27, 2002
    Messages:
    643
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    21
    Location:
    By, and sometimes in, the sea
    Ratings:
    +2
    Sorry to be equivocal but the short answer is - it depends....

    The long answer is far too long for a post here.

    For the moment, be guided by your ears. If it sounds better with the speakers set to big, then fine. If it sounds better the other way then - make up your own mind.
     
  6. eddiea

    eddiea
    Guest

    Products Owned:
    0
    Products Wanted:
    0
    Ratings:
    +0
    I also read this and tried it, my setup did not sound right at all, so it's now back to using small speaker settings.

    Be guided by what sounds best to you.

    Regards

    Eddie:D
     
  7. dUnKle

    dUnKle
    Well-known Member

    Joined:
    Jan 1, 2001
    Messages:
    13,364
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    136
    Ratings:
    +1,319
    No, no, no

    Not again, please ;)

    In the end I went for large, but only for the fronts, and then thats with the subwoofer connected by both speaker and phono connections.

    It does now sound miles better than ever before, with tight controlled bass and clear sound all round, but boy did it take some getting correct !
     
  8. Ian J

    Ian J
    Banned

    Joined:
    Aug 6, 2001
    Messages:
    25,529
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    166
    Ratings:
    +4,906
    The amps tested were :-

    Arcam AVR200
    Denon 3802
    HK 5500
    Onkyo DS696
    Sony VA555ES

    The Arcam and HK came out worst (quote "firmly slammed into overload") whilst Onkyo managed much better with the remaining two in between.

    The above comments are in relation to bass management only.
     
  9. Ian J

    Ian J
    Banned

    Joined:
    Aug 6, 2001
    Messages:
    25,529
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    166
    Ratings:
    +4,906
    Speakers were Mission 782, 78c and 78ds plus Acoustic Energy Compact for extra surrounds.

    Subwoofer was REL Q150E
     
  10. Reiner

    Reiner
    Active Member

    Joined:
    Jul 16, 2000
    Messages:
    3,315
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    61
    Location:
    Germany
    Ratings:
    +13
    Now, variable low-pass filter adjustment (as used by most good subs) is a great feature that helps to blend a given sub to the rest of the speaker system. By using a "pre-filtered" signal (in most case's setting speakers to small on your amp/processor) this rather defeats the purpose of this great feature.
    However, most higher end amps/processors allow the defeat of built in low pass filters 'or' select the frequency to allow for a better 'meeting' with the sub.


    IMHO that's not necessary when using the line level connection only as it is common / recommended in AV setups. As the amp/receiver will take care of the bass management the x-over at the subwoofer can be set to maximum and thus it actually not needed. Not a true bypass but no need to "better meeting" either.
     
  11. garmtz

    garmtz
    Standard Member

    Joined:
    Mar 27, 2002
    Messages:
    305
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    21
    Location:
    Netherlands
    Ratings:
    +1
    These are complicated matters and bass-management is the most misunderstood subject of home cinema. I have not yet read the article, but want to mutter some personal opinions/ideas.

    Bass management has been invented to prevent small speakers being overdriven by strong bass signals, therefore reducing the load on the amplifier and reducing distortion of the speaker (less strain on the mostly small woofers of most speakers and less intermodulation distorion of the speaker, which makes for a cleaner midrange with better imaging).

    On the other hand, by filtering the bass from the speakers, the load on the subwoofer is increased. The maximum output level for the normal speakers at THX reference level is 105 dB sound pressure. For the sub, this is 115 dB. Not many people listen at reference level, but at least 10 dB lower, so the subwoofer needs to have clean output up to 105 dB. Adding the rerouted bass from the other speakers adds another couple of db's (could be 6 dB), making the total output 111 dB. This already is pretty extreme and VERY few subwoofers can handle this.

    I have some experience with REL subs and found that they can overload pretty quickly in larger rooms.

    What I think Hifi Choice has done is to measure the performance limits of the used REL Q150E subwoofer, NOT the pros and cons of bass management. If they had used an ultra high-performance sub like the Paradigm Servo-15, a large Velodyne, Mirage or Sunfire, the results would probably have been VERY different. The reason the HK and Arcam were worst probably have to do with their crossover frequency being higher than the other amps, which puts even more strain on the subwoofer.

    I my own experience, my speakers sound MUCH better with a small setting: less bass distortion, better imaging, improved maximum output etc. etc. I must add that in my processor, the LFE is handled by a separate subwoofer (BagEnd Infrasub-18) and all rerouted bass from the speakers is handled by two REL Q100E subwoofers. All crossovers are set at maximum, filtering is handled by the processor (60 Hz for fronts and stereo side subs, 120 Hz for center, 90 Hz for sides, 70 Hz for rears).

    In the end, it's another article written from the wrong perspective and with a lot of preconceptions and assumptions. It is difficult to know all parameters in such an ordeal.
     
  12. Nic Rhodes

    Nic Rhodes
    Well-known Member

    Joined:
    Mar 23, 2001
    Messages:
    17,133
    Products Owned:
    0
    Products Wanted:
    1
    Trophy Points:
    133
    Location:
    Cumbria
    Ratings:
    +1,277
    Well many of the keen eyed readers here might have noticed I have stayed out of the Large / small debate. This was deliberate. It was a debate I had mixed feelings over having achieved completely different results using different receivers using the same amp and speakers. This distortion / level issue discussed in the article was the reason why. It many ways it reminds me of the blanket recommendations by people of RGB Scart over all over DVD video interfaces. Those who took the effort to experiment occasionally found SVideo better and invariably component always is. The large / small debate is similar. It is all down to quality of engineering (spot the engineer!!)

    Bass management is now becoming a key issue. People require flexibility in what they do. SACD players have been hampered in their sales because of no bass management. This would normally be done digitally in the processor but Sony in their wisdom have yet to release a player with a digital output (for SACD anyway). Some of the competition have implemented IEE1394 interfaces. Stung by criticisms of lack of bass management Sony have added this to their latest SACD machines. However as DSD is a 'baby' in the technology front, people are 'unsure' how to process this data digitally. Sony’s current solution is to convert to PCM for bass management. This is the whole reason Sony was 'marketing' SACD to get away from PCM and make things less complicated! It might also explain why people like me are less than happy with budget / mid priced SACD players.

    Now lets look at these processors and receivers who do 'bass management' in their respective boxes.

    Bass management is done by the DSP in these boxes. These need to be controlled by some sophisticated computer software. One box low/mid market solutions use one chip-integrated solutions. These generally perform poorly in this regard. Jitter and all that (>1000ps) again. When it come to bass management, quite frankly they perform very poorly I have therefore all been happier leaving everything set as large unless you have a real tiny speaker. Don't forget for stereo we have fed full range signal to some really small speaker with limited / no problems.

    I have argued repeatedly on this board for low distortion devices, both amplifiers and subwoofers in particular. I have argued the case for low distortion 'servo' sub woofers. I am consistent amazed how many glowing reports of bass quality come through here, despite the fact the receiver could be producing 10% distortion and the sub can add another 30%. This is common but there are many here who write glowing reports on kit performing like this. It is clearly audible and sounds dreadful.

    Proper ENGINEERED DSP chipsets perform rather better in this respect. In this context I mean the top Motorola and more specifically SHARCs used now by Lexicon, Tag and Denon. These are all high quality 'generic' DSP processors but are well controlled by software. They perform very much better in this respect and do keep distortion low. Bass management works much better now and speakers can be set confidently to large / small.

    Even with these better performing chipsets, much is down to the quality of software programming. This is where Meridian, Lexicon, Theta and Tag come into their own. Software on these units has taken years to evolve, ironing out all these bugs. Your mass-market solution you just get it with all the 'bugs'.

    I first came across this when playing with my then new Servo15s. It was an 'open' window to allow me to see where the distortion was going. The combination with Tag stuff is awesome, with mid market receivers, the low distortion Servo15s shows high levels of distortion with some models.

    There are two ways around this.

    1. is to set all speakers to large and sub woofer set ON.

    2. this is what I think will be the best combination but is expensive. Use analogue filter for sub woofers on L and R channels. This requires two subs on the main channels. Then add a third sub of the low frequency effects channel. Avoid all bass management.

    I suppose there is a third, get a receiver / processor that works in the fist place!!. The old Arcam 100 got around this as well by not having bass management (!) but I wasn’t a fan of this Arcam product.

    Do we hear differences when we change from small to large with most of these receivers. Yes we do. They are distortion and level differences. Are these better? Not in my book but as Stuart Robinson has repeatedly pointed out NO one is willing to ABX listening tests under controlled conditions with levels matched to 0.1db. Try arguing anything else to AES members and they will laugh you out of the room.

    On this subject anyone interested in taking up Stuarts R cable challenge or are you content in spending foolish amounts of money on cables?

    We assume as we are hearing a difference and there is much more bass that previously (due to level and distortion) it must be better. It ain't, at least in many situations particularly with distortion / level problems. I just wish I had $1 for every claim of better sound. I would be a happy man with those Velodyne 18 units, both of them!

    My real concern is over sub sat systems and how they perform with poorly designed processors. There is a current tendency to use digital rather than analogue filtering. This may be fine with a MC12 but a cheap HK?

    And people wonder why I so concerned about technical engineering / specs?

    Don't forget the distortion we are dealing with come from the digital processing stages and has nothing to do with analogue distortion / loads etc. [Back to my digital done correctly stance I have argued for years]. Numbers or % in the digital distortion game can escalate VERY quickly. (early CD players often had distortions levels at >100% at low signal levels, the days when we didn't know about digital and CD players sounded awful).

    If you are going to try this you must do it with a SPL meter and make sure you are re calibrated correctly, otherwise it is meaningless. Don't forget all digital equipment is not created equal, it isn't an issue with some but is with others. It is down to QUALITY ENGINEERING or not as the case may be.

    A few more general thoughts:

    What about sub / sat systems now very popular. It might be beneficial to leave all the bass management to the 'old' analogue stages. Doesn't quite tally with the current wisdom does it! My advice is to suck it and see but don't expect to set stuff up in minutes. This takes time (hours / days) this time is time well spent. Doesn't quite tally with a 30-min demo in a shop syndrome definitely proving one way is better than another.

    THX does loads of bass management along with several other very useful features. The current trend is to knock THX particularly with music (where the original spec was only ever designed for films, and most of them were PL only and not discrete DD / DTS). I am not in this camp. I think THX was a valuable addition to AV though I am glad they have addressed issues with discrete formats and music in THX2. No one liked spending those expensive THX fees just to have their kit 'checked out'. Well guys you now know why THX insisted on it. THX was a guarantee that everything performed to their (high) standards. Yes there is good performing kit that wasn't THX but there are no THX certified kit I know of with distortion problems (cue a flurry of PMs). THX wouldn't allow it. THX approved hardware is a minimum guarantee of quality (even if their videos aren’t!).

    The Outlaw ICBM might be another potential solution.

    Found a few minutes away from young children!!, had to go to London for 3 days!
     
  13. garmtz

    garmtz
    Standard Member

    Joined:
    Mar 27, 2002
    Messages:
    305
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    21
    Location:
    Netherlands
    Ratings:
    +1
    I certainly do! :) (but hey, I am a single guy, it figures...)
     
  14. stranger

    stranger
    Guest

    Products Owned:
    0
    Products Wanted:
    0
    Ratings:
    +0
    this is what i have done and i only wish i had room to put subs on the remaining chanels.
     
  15. garmtz

    garmtz
    Standard Member

    Joined:
    Mar 27, 2002
    Messages:
    305
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    21
    Location:
    Netherlands
    Ratings:
    +1
    I only use 2 subs for rerouted bass (placed below the side channels) and I think this is enough. The bass management of the Lexicon MC-12 is pretty transparent and Lexicon believes in routing bass to the sides of the listener anyway. Their "bass enhance" feature, which I use in all modes where this function is available, is based on this. It makes the sound much more fluid and spacious.

    It is not really the case what the speakers can reproduce or that they will be overloaded, but the issue is what smaller speakers can reproduce cleanly. In most cases, sound quality of a smaller speaker (able to reproduce frequencies without any attenuation down to about 50 Hz) will improve when used with a crossover @ 60-80 Hz. Why send bass to a speaker which is unable to reproduce these frequencies? It will only add distortion.
    With careful calibration/setup and the use of good quality subwoofers, most of these problems can be solved.
     
  16. Plump

    Plump
    Standard Member

    Joined:
    Feb 22, 2002
    Messages:
    636
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    19
    Location:
    under the bridge
    Ratings:
    +0
    Just to add one here,
    this one is very, I mean VERY, simple.
    If you have awfull lot of time to loose try experimenting, but be aware you might end wrong.
    If not, follow advice I got from one KEF engineer three years ago on Sound'n'Vision : "If you have decent speakers, set 'em all big and you won't end wrong - If you have computer speakers you're in the wrong place here!"

    Cheers
     
  17. Charlie Whitehouse

    Charlie Whitehouse
    Guest

    Products Owned:
    0
    Products Wanted:
    0
    Ratings:
    +0
    Having now read this article, I have a few comments to add to the debate here.

    Firstly, Paul Miller is a well-respected, nay guru, on testing of digital equipment. However, this test leads to more questions than answers.

    Did nobody notice that they managed to print the distortion graph again instead of the subwoofer level?? This sort of sloppy proof-reading leads one to wonder how much store you can set by some of the measurements that get printed in magazine reviews in general!!

    From my reading of the test, he was testing the signal at the sub-out terminals; no speakers or subs would have been necessary in this case so the question of what reference gear was used in the test becomes irrelevant.

    As to the audibility of distortion and since we're singing the virtues of ABX testing, you should read AES Pre-Print 1771 (H-4) High Resolution Subjective Testing Using A Double Blind Comparator. There is an interesting section towards the end which talks about some real results and applications of the ABX Comparator. For example, when used to test pre-amplifiers or amplifiers where the responses are matched, and I quote:

    To find out what amount of distortion was audible, a distortion generator was developed. Nicknamed the "Grunge Box", it generates even order harmonic components that are independent of level. The RMS output also remains fairly consistent as %THD is varied. Distortion can be heard on high or low level music passages and there are no level set problems. No two real world non-linearities are the same and the "Grunge Box" is yet another one, but its sound is somehow "typical" and useful.

    Three percent is the best done so far, but with carefully selected material (flute solo etc.) two or one percent might be possible. One percent is easy with sine waves.

    The worst of medium quality electronics only approaches .5% THD mid band when driven near clipping. It is not surprising then, that no differences were heard.


    In this context, the increase in distortion observed by Paul Miller when going from 5.0 to 5.1 DD bitstreams with the Onkyo and Denon where distortion remains below 0.1% is unlikely to be audible in practice. With the Sony, HK and Arcam where it zooms up to 8-12% it most definitely will be audible.

    I'm content to spend foolish amounts on cables, but I am not unwilling to try a blind test. I suspect that many other audiophiles are not actually frightened of such a test - it's the practicalities of doing auditions of stuff on this basis that causes the problem. I doubt that many dealers would lay on such a test. Perhaps between the members of this forum we could accomplish such a thing? I for one would be more than happy to participate. So how's about it, chaps? :rolleyes:
     
  18. garmtz

    garmtz
    Standard Member

    Joined:
    Mar 27, 2002
    Messages:
    305
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    21
    Location:
    Netherlands
    Ratings:
    +1

    To speak on my own behalf: I have not yet read the article and I probably should. My comments were based on experience and theoretical knowledge acquired in recent years. I will read the article and try to give more comments or to negate my own posts... :)

    The problem however is that distortion figures and theoretical explanation most of the time don't tell the whole story. I also choose my cables very carefully, because although they don't distort the sound "to an audible level", I still find that they can make or break a system. No need for me to be convinced by an ABX listening test. But this already seems to be a controversial statement.
     
  19. Mikael

    Mikael
    Guest

    Products Owned:
    0
    Products Wanted:
    0
    Ratings:
    +0
    There is also another purpose of bass management, apart from avoiding damage to your speakers. If the signal contains audio down to 30 Hz or so, as most modern soundtracks tend to, it needs to be redirected to a speaker that can actually reproduce it, otherwise it won't be heard, right? That was one of the problems with the first issue of Jurassic Park dts, much of the bass was in the main channels, and people with incorrect bass management setups didn't hear it, and said the soundtrack was mixed wrong, which in fact it wasn't.

    So, depending on how the sound engineer mixed the soundtrack, lots of bass may exist in channels other that LFE, and if this is sent to speakers unable to reproduce it, it will be lost...
     
  20. Nic Rhodes

    Nic Rhodes
    Well-known Member

    Joined:
    Mar 23, 2001
    Messages:
    17,133
    Products Owned:
    0
    Products Wanted:
    1
    Trophy Points:
    133
    Location:
    Cumbria
    Ratings:
    +1,277
    Another thought

    What damages speakes more, low frequency output or distortion by poor processing?

    Or is the overload just more analogue distortion to add to the digitally derived distortion.
     
  21. garmtz

    garmtz
    Standard Member

    Joined:
    Mar 27, 2002
    Messages:
    305
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    21
    Location:
    Netherlands
    Ratings:
    +1
    I think distortion within the rated frequency response and power handling of the speaker is less damaging than a speaker trying to reproduce a signal that the cone cannot physically reproduce cleanly. Especially ported speakers are prone to very high levels of distortion (this does not always mean damage) with very low bass signals, because the cone is not particularly well damped. A closed box is much less prone to distorting with low bass.

    An exception must be made for the tweeter of most speakers. Most are easily damaged by clipping, but I doubt if this is the kind of distortion that occurs when using a crossover, because most of the distortion will be at low frequencies.
     
  22. Charlie Whitehouse

    Charlie Whitehouse
    Guest

    Products Owned:
    0
    Products Wanted:
    0
    Ratings:
    +0
    This is what I find unsatisfactory about Paul Miller's article. It doesn't attempt to define what the nature of the distortion is. Is it harmonic, phase, intermodulation, or whatever? More questions than answers as I said....

    It is my understanding that most speaker damage tends to occur due to amplifier clipping. This is due to the fact that the ensuing effective 'square-wave' envelope, when subject to fourier analysis yields a whole host of high frequency components. It is these when presented at high-level that have the effect of frying tweeters. Damage results from the heating effect of these HF currents on the tweeter voice coils.

    Distortion, per se, is not going to damage speakers. It is simply the presence of signal components not in the original. Provided the operating envelope of the speaker is not exceeded, no damage should occur. However, driving a woofer with a high-power LF signal may cause such cone excursion that physical damage occurs to the cone or surround.

    To answer Nic's question, I think that feeding an excessive LF signal to your speakers is more likely to cause damage than a bit of distortion.

    That's OK with me. I seem to remember defending Airwise over SMR's review of the Theta Dreadnaught without ever hearing it! None of us is perfect! :D

    Well this is kind of my own position, but I keep getting told that differences are not audible under blind testing conditions and I would like to think that I am open-minded enough to try it. Seeing (or hearing) is believing as they say!

    Well I don't remember promising to organise this thing. 'Participate In' was the phrase that I used, I think. However, if there is sufficient interest I don't mind trying to organise something. Also, I don't mind loaning equipment, cables, my time etc. However, I don't really have the room to accommodate a large party here, so if someone else is prepared to suggest an alternate venue, that would help.

    If anyone else is interested in this, let me know and perhaps we should start a separate thread to discuss requirements/arrangements. :)
     
  23. ads1

    ads1
    Active Member

    Joined:
    Mar 13, 2002
    Messages:
    71
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    8
    Location:
    Cumbria
    Ratings:
    +15
    Although not true ABX, there is an interesting experiment underway on the forum of another hi fi mag, in which one member has made up cables from differing quality materials, masked them and these are then circulated around the participants. It might be fun to try take part.

    ads1

    PS just checked. the board is down for the next month :-(
     
  24. Charlie Whitehouse

    Charlie Whitehouse
    Guest

    Products Owned:
    0
    Products Wanted:
    0
    Ratings:
    +0
    Since this discussion seems to be going all over the place, let's go back to what the article ACTUALLY says and argue from there.

    The first graph presented compares distortion figures measured at the Sub-out terminals in two scenarios. In BOTH scenarios, all 5 channels are set to 'small', therefore crossed-over bass from these 5 channels is presented at the Sub-out terminal. In the first instance, there is no other LFE signal present, in the second instance there is.

    The second graph SHOULD have shown the difference in the signal level of the crossed-over bass present in the Sub-out channel in the same two scenarios. However, due to typographical error, this graph is not printed, the distortion graph is printed again. With this error, it is not really possible to deduce too much about the relative performance of the amplifiers under test, so let's ignore this finding, apart from noting that significant level changes are present.

    Of course, the test 'subject' is crucial. My reference to 'reference gear' was in relation to discussion earlier in the thread by Chip and Ian as to which speakers and subs had been used in the test. In the context of the 'Group' listening tests of these amplifiers presented earlier in the magazine, the choice of speakers and subs used are of course highly relevant. However LAB bench tests of amplifiers in my experience rarely test them with real speakers attached - too many variables introduced. Instead they have them driving dummy standard loads which are usually almost pure resistance. Another reason why such test results tell you almost nothing about how an amplifier will perform when driving a real-world reactive load like a loudspeaker!

    In this context, while Eric is correct that the 'cascading' of crossovers in amplifier and sub is almost bound to be detrimental, it does not explain the measurements here.

    If you examine the THD figures presented in the test for the scenario with no LFE, they closely mirror the THD figures for the respective amplifiers, at moderate output levels, presented in the earlier Group Tests. I.E., the redirected bass is no more distorted by this process than any of the other channels. The real shocker is not the quality of the actual crossover function, it is the inability of some of these amplifiers to COMBINE the re-directed bass with any LFE, should it be present.

    What surprised me was the huge variation between some amplifier's abilities to do the 'combination' compared with others. Perhaps, I have been spoilt with my Casablanca but I had always assumed that the digital processes of 'standard' surround decoding and bass-management were well-understood, off-the-shelf algorithms that could be implemented by mass-market manufacturers as well as the specialists. Although, the mass-market would lack the flexibility of the 'soft-programmability' of the high-end boys and their proprietary surround decoding algorithms, by employing off-the-shelf standard chipsets they would achieve reasonable performance with high reliability. Furthermore, I had assumed that if you coupled the digital processing of a mass-market amplifier with some decent power-amps you would have a combo that would compete relatively well with the best-of-the-best pre/pros. It would seem it ain't that simple!! :(

    Onkyo and Denon seem to manage the feat reasonably well while Sony, HK and Arcam all fail miserably in this test. Perhaps HCC might consider applying similar tests to AV amplifiers and processors that they test in future. They seem to offer a much larger variance in performance against which one could make a choice.

    So what do you do with this information? Are we saying that because some manufacturers cannot do bass-management properly we must all set our speakers to 'large'? If we do, then bass will be directed to speakers that cannot reproduce it. While this may or may not result in audible distortion to other parts of the signal, it will place additional strain on the power amplifier which will attempt to reproduce those bass frequencies which we will never hear. That energy has to go somewhere too. This is bound to limit the dynamic headroom of the system.

    We will also be missing out on a substantial amount of bass information. For music purposes, OK we probably wouldn't miss it, but for AV use, that bass is integral to the enjoyment of the experience.

    Take for example U571. As I showed in a thread last year, much of the bass in the depth-charging scenes is present in the main channels. If you are going to set everything to 'large' you are surely going to be missing out on much of the 'thrill' unless you have some huge floorstanders!

    Yes, but that sort of thing happens all the time. Even if you keep channels discrete right up to the individual loudspeaker, this sort of combination occurs within your room. It is in the very nature of polyphonic harmony and is really what music is all about. Subtle phase differences are what allow our ears to locate individual sounds in space.

    This may or may not be true. Have you done a blind test to prove it? :devil: Only kidding! :D It is worth pointing out that these frequency cut-offs for cross-overs are not brick-walls so that 'everything' below the frequency goes to the sub. There is a gradual 'slope' involved. In fact, there are two slopes - high-pass (going to the main channel speaker) and low-pass (going to the sub). In the overlap zone of these two, any differences in the responses of the two signal paths result in peaks or troughs in the overall response. There will usually be phase differences between the bass signals present in the high-pass and low-pass areas resulting in 'smearing'. This would be further compounded if the crossover in the sub were in use as well. This is what I assume you mean by 'cascading' problems.

    On the subjectivist/objectivist thing, let me say that I am strongly in the subjectivist corner. I tend to trust my ears. However, I recognise that if something is audible, then it should be measurable and repeatable. In this context I can see no objection to double blind testing as an ideal. I would love to prove that cable differences ARE audible, hence my interest in participating in a blind test. The risk that I might find they are NOT is one I am prepared to take. :rolleyes:
     
  25. Jeff

    Jeff
    Well-known Member

    Joined:
    Jul 13, 2000
    Messages:
    5,489
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    106
    Location:
    Basingstoke
    Ratings:
    +256
    Not much consideration is given to the load on the amp when setting speakers to large, isn't this an issue? On the other hand if you are going to set all the speakers to small you need a sub with sufficient power to take the load, is 100-200W enough?
     
  26. garmtz

    garmtz
    Standard Member

    Joined:
    Mar 27, 2002
    Messages:
    305
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    21
    Location:
    Netherlands
    Ratings:
    +1
    Exactly Jeff! The tests that Hifi Choice performed never took the load on the power amp into consideration. This, combined with distortion from drivers not able to reproduce very low bass can in fact dwarf the distortions measured at line level! I know this was not the issue of the test, but you cannot ignore the practical considerations when doing such an investigation.

    All my points above still hold. I also stongly agree with Charlie's opinion.

    How much power you would need in the subwoofer would be dependant on the chosen crossover frequency, the amount of speakers crossed over to the sub, the efficiency of the conus used and the amount of eq, the size of the enclosure, the loading of the subwoofer cone, the size of the room, the placement of the sub, the furniture in the room, the desired maximum sound pressure levels... :eek: Not easy to put a figure on this... :)
     
  27. stranger

    stranger
    Guest

    Products Owned:
    0
    Products Wanted:
    0
    Ratings:
    +0
    the tests also don't state what the amps' bass m. crossovers were set at for any of the tests. i would think this would vary widely, the sony for instance has variable crossover settings from 40hz. to 200hz. as in any short reviews and tests like this they seem to leave more questions than answers.
     
  28. Charlie Whitehouse

    Charlie Whitehouse
    Guest

    Products Owned:
    0
    Products Wanted:
    0
    Ratings:
    +0
    Ron,

    If you want to know what an ABX comparator looks like, then look here: ABX Comparator . There is some useful information on test results undertaken with this equipment, plus information on the latest PCABX virtual equivalent.

    I don't think the ABX Comparators are manufactured any more so borrowing one is probably the only answer now. In any case, I don't believe the use of such a device is mandatory for blind testing, it just makes things easier.
    I'm not sure whether Chip is advocating a shootout between different processors/amplifiers, but I thought that blind testing was introduced into this thread by Nic in the context of SMR's cable challenge. As far as I recall this was a £1000 ish silver interconnect that if anyone could tell from another cable in a blind test would be presented as a prize - no winners yet!

    In terms of what would be interesting to test, my suggestions would be:
    Optical vs Coax Digital
    Low vs high quality interconnects
    Normal vs expensive power cables
    Single vs Bi-wire speaker cables
    The results of such a test should be pretty interesting to all here I would have thought.
    Any other suggestions?
     
  29. stranger

    stranger
    Guest

    Products Owned:
    0
    Products Wanted:
    0
    Ratings:
    +0
    i am one of those who believes that cables do make a difference and have proved it to my satisfaction (personal). at one time i tried this simple test sugested by russ andrews. i had received some vs speaker cable from him one was labelled for directionality-one wasn't.using one chanel only, conect cable up and play some music with treble content- reverse cable and try again, i heard a definite difference try it -you dont have to drag your cables out just disconnect and try a spare piece. think i meandered on this a bit but i think my point is clear.
     
  30. Nic Rhodes

    Nic Rhodes
    Well-known Member

    Joined:
    Mar 23, 2001
    Messages:
    17,133
    Products Owned:
    0
    Products Wanted:
    1
    Trophy Points:
    133
    Location:
    Cumbria
    Ratings:
    +1,277
    The measured distortion was 12% ish without any load. You are saying that the 'analogue' load distortion is orders of magnitude higher then some people are listening to seriously distorted kit out there. As I have said repeatedly before distortion on subs can easily hit 30% in addition to the 'digital' distortion unless you are a servo guy.

    Why do people tollerate this?

    Charlie, excellent post I think people were missing the issue somewhat, probably due to not having read the article!:confused:
     
Thread Status:
Not open for further replies.

Share This Page

Loading...